Computational design of synthetic regulatory networks from a genetic library to characterize the designability of dynamical behaviors

The engineering of synthetic gene networks has\udmostly relied on the assembly of few characterized\udregulatory elements using rational design principles.\udIt is of outmost importance to analyze the scalability\udand limits of such a design workflow. To analyze the\uddesign capabilities of libraries of regulatory\udelements, we have developed the first automated\uddesign approach that combines such elements to\udsearch the genotype space associated to a given\udphenotypic behavior. Herein, we calculated the\uddesignability of dynamical functions obtained from\udcircuits assembled with a given genetic library.\udBy designing circuits working as amplitude filters,\udpulse counters and oscillators, we could infer new\udmechanisms for such behaviors. We also highlighted\udthe hierarchical design and the optimization\udof the interface between devices. We dissected the\udfunctional diversity of a constrained library and we\udfound that even such libraries can provide a rich\udvariety of behaviors. We also found that intrinsic\udnoise slightly reduces the designability of digital\udcircuits, but it increases the designability of oscillators.\udFinally, we analyzed the robust design as\uda strategy to counteract the evolvability and noise\udin gene expression of the engineered circuits within\uda cellular background, obtaining mechanisms for robustness\udthrough non-linear negative feedback\udloopsFunding for open access charge: Generalitat Valenciana\ud(BFPI-2007-160); HPC-Europa programme (RII3-CT-\ud2003-506079); Spanish Ministry of Education and\udScience (TIN-2006-12860); Structural Funds ERDF;FP6-NEST 043340 (BioModularH2); FP7-ICT-043338\ud(BACTOCOM); FP7-ICT-265505 (CADMAD); ATIGE\udGenopole/UEVE (A3405); Fondation pour la Recherche\udMedicale.Peer reviewe